[
https://issues.apache.org/jira/browse/FLEX-34648?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14581823#comment-14581823
]
Robert Hunt commented on FLEX-34648:
------------------------------------
I'm not sure if this is the same issue but we discovered a memory leak in our
application related to a huge number of AsyncMessages being accumulated over
time. We identied the issue in production by enabling "heap dump on crash" (in
Tomcat) and then inspecting the heap dump using a profiler. We were later able
to spot the issue as it was hapenning when we knew where to look.
What we found was that FlexSessions were accumulating a number of invalidated
FlexClients. These FlexClients had outboundQueues full of undelivered
AsyncMessages. Once a FlexClient is invalidated it no longer accepts messages
into it's queue but it can take a few minutes for it to become invalidated (in
our configuration) which meant they sometimes accumulated a fair amount of
messages before becoming invalidated.
You'd expect invalidated FlexClients to be garbage collected but we found they
were not, this was due to a strong reference being held in this bit of code:
https://github.com/apache/flex-blazeds/blob/master/modules/core/src/flex/messaging/endpoints/BaseStreamingHTTPEndpoint.java#L1105
Once it drops into the handleFlexClientStreamingOpenRequest method it ends up
in a "wait-notify loop" and due to the strong reference to the FlexClients via
the collection they can't be marked for garbage collection.
Our fix involves subclassing BaseStreamingHTTPEndpoint and overidding the
serviceStreamingRequest method. We inserted the statement
"flexClients.clear();" after this line:
https://github.com/apache/flex-blazeds/blob/master/modules/core/src/flex/messaging/endpoints/BaseStreamingHTTPEndpoint.java#L1115
> [BLAZEDS]Memory Leak occurred in AsyncMessage when sending alot of
> -------------------------------------------------------------------
>
> Key: FLEX-34648
> URL: https://issues.apache.org/jira/browse/FLEX-34648
> Project: Apache Flex
> Issue Type: Bug
> Components: BlazeDS
> Affects Versions: BlazeDS 4.7
> Reporter: [email protected]
> Priority: Critical
>
> a memory leak occurred when sending alot of AsyncMessage through BLAZEDS in a
> real time systems which is heavilly using messaging however we are increasing
> the jvm heap size to 4 GB 80% of the size is occupied by AsyncMessage
> objects this is very clear from the generated heap dump.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
